On Tue, 25 Mar 2014 00:18:54 -0400
"J. Bruce Fields" <bfields@xxxxxxxxxxxx> wrote:
> On Mon, Mar 24, 2014 at 05:57:11PM -0700, Jeffrey Layton wrote:
> > On Mon, 24 Mar 2014 20:20:47 -0400
> > "J. Bruce Fields" <bfields@xxxxxxxxxxxx> wrote:
> >
> > > On Wed, Mar 19, 2014 at 04:45:54PM -0400, Jeff Layton wrote:
> > > > In a later patch, we'll be adding a new type of lock that's owned by
> > > > the struct file instead of the files_struct. Those sorts of locks
> > > > will be flagged with a new IS_FILE_PVT flag.
> > > >
> > > > Add a "P" suffix to the POSIX lock output in /proc/locks for locks
> > > > that have FL_FILE_PVT set to distinguish them from "classic" POSIX
> > > > locks.
> > >
> > > ACK, though I'm curious:
> > >
> > > > Signed-off-by: Jeff Layton <jlayton@xxxxxxxxxx>
> > > > ---
> > > > fs/locks.c | 9 +++++++--
> > > > include/linux/fs.h | 1 +
> > > > 2 files changed, 8 insertions(+), 2 deletions(-)
> > > >
> > > > diff --git a/fs/locks.c b/fs/locks.c
> > > > index 4d4e790150e0..f1b9b149dd39 100644
> > > > --- a/fs/locks.c
> > > > +++ b/fs/locks.c
> > > > @@ -135,6 +135,7 @@
> > > > #define IS_POSIX(fl) (fl->fl_flags & FL_POSIX)
> > > > #define IS_FLOCK(fl) (fl->fl_flags & FL_FLOCK)
> > > > #define IS_LEASE(fl) (fl->fl_flags & (FL_LEASE|FL_DELEG))
> > > > +#define IS_FILE_PVT(fl) (fl->fl_flags & FL_FILE_PVT)
> > > >
> > > > static bool lease_breaking(struct file_lock *fl)
> > > > {
> > > > @@ -2313,8 +2314,12 @@ static void lock_get_status(struct seq_file
> > > > *f, struct file_lock *fl,
> > > > seq_printf(f, "%lld:%s ", id, pfx);
> > > > if (IS_POSIX(fl)) {
> > > > - seq_printf(f, "%6s %s ",
> > > > - (fl->fl_flags & FL_ACCESS) ?
> > > > "ACCESS" : "POSIX ",
> > > > + if (fl->fl_flags & FL_ACCESS)
> > > > + seq_printf(f, "ACCESS");
> > >
> > > Are FL_ACCESS locks ever put on any global list? Wouldn't it be a bug
> > > to run across one here?
> > >
> > > --b.
> > >
> >
> > Yes, that can happen (and yes, that surprised me too)
> >
> > locks_mandatory_area can send down FL_ACCESS|FL_POSIX|FL_SLEEP
> >
> > ...which makes sense given the use-case there but I do wonder if there
> > might be a better way...
>
> Got it!
>
> Yeah, that's weird, but I guess it's right.
>
> --b.
Thanks for the ACK, but I think I'd like to modify this patch a bit.
Since POSIX hasn't accepted this yet, calling these POSIXP locks is
somewhat of a misnomer. How about this patch instead?
-----------------------[snip]------------------------
[PATCH] locks: make /proc/locks show IS_FILE_PVT locks as type "FLPVT"
In a later patch, we'll be adding a new type of lock that's owned by
the struct file instead of the files_struct. Those sorts of locks
will be flagged with a new FL_FILE_PVT flag.
Report these types of locks as "FLPVT" in /proc/locks to distinguish
them from "classic" POSIX locks.
Signed-off-by: Jeff Layton <jlayton@xxxxxxxxxx>
---
fs/locks.c | 11 +++++++++--
include/linux/fs.h | 1 +
2 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/fs/locks.c b/fs/locks.c
index 4d4e790150e0..0e1f0df8de12 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -135,6 +135,7 @@
#define IS_POSIX(fl) (fl->fl_flags & FL_POSIX)
#define IS_FLOCK(fl) (fl->fl_flags & FL_FLOCK)
#define IS_LEASE(fl) (fl->fl_flags & (FL_LEASE|FL_DELEG))
+#define IS_FILE_PVT(fl) (fl->fl_flags & FL_FILE_PVT)
static bool lease_breaking(struct file_lock *fl)
{
@@ -2313,8 +2314,14 @@ static void lock_get_status(struct seq_file *f, struct file_lock *fl,
seq_printf(f, "%lld:%s ", id, pfx);
if (IS_POSIX(fl)) {
- seq_printf(f, "%6s %s ",
- (fl->fl_flags & FL_ACCESS) ? "ACCESS" : "POSIX ",
+ if (fl->fl_flags & FL_ACCESS)
+ seq_printf(f, "ACCESS");
+ else if (IS_FILE_PVT(fl))
+ seq_printf(f, "FLPVT ");
+ else
+ seq_printf(f, "POSIX ");
+
+ seq_printf(f, " %s ",
(inode == NULL) ? "*NOINODE*" :
mandatory_lock(inode) ? "MANDATORY" : "ADVISORY ");
} else if (IS_FLOCK(fl)) {
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 7527d96913d3..5ddeb8de5e77 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -888,6 +888,7 @@ static inline int file_check_writeable(struct file *filp)
#define FL_SLEEP 128 /* A blocking lock */
#define FL_DOWNGRADE_PENDING 256 /* Lease is being downgraded */
#define FL_UNLOCK_PENDING 512 /* Lease is being broken */
+#define FL_FILE_PVT 1024 /* lock is private to the file */
/*
* Special return value from posix_lock_file() and vfs_lock_file() for
--
1.9.0
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
